We validated the therapeutic potential of 8 genes (COP1, CSN5, HDAC2, R1 (RAM2), RBX1, WEE1, FOXM1 and CDK4), all with high hazard ratio for survival. For each gene, three specific siRNAs encapsulated in stable-nucleic-acid-lipid-particle (SNALP) were designed and tested for growth inhibition in two HCC cell lines (Huh7 and HepG2) as determined by MTT and TUNEL assays. Reduction in target RNA was confirmed by qRT-PCR analysis. The siRNA targeting of COP1, CSN5 and HDAC2 genes in the HCC cells restored the levels of p53 and its target p21, suggesting that siRNA-induced apoptosis was p53-dependent. siRNAs showing therapeutic activity in cell culture were further evaluated in vivo using a luciferase-based reporter cell line Huh7luc+ and an orthotopic xenograft model. Chemically modified siRNAs were systemically delivered to liver through tail vein injections. SNALP was selected as an optimal carrier for siRNA delivery into the mouse because of the enhanced stability, lack of toxicity and smaller effective dose. Compared to the SNALP control siRNA treatment, 3-4 injections of 2 mg/kg SNALP containing COP1, CSN5, HDAC2, RBX1, WEE1 and CDK4 siRNAs, effectively suppressed HCC cell growth in vivo. The treatment with SNALP FOXM1 siRNA as well as R1 siRNA had no effect on HCC cell growth. Of the 6 genes validated in vitro and in vivo, silencing of CSN5 and WEE1 caused a two-fold reduction in the proportion of side population both in Huh7 and HepG2, suggesting that these siRNAs may be effective in anti-cancer stem cell therapy. Together, these results indicate that COP1, CSN5, HDAC2, RBX1, WEE1 and CDK4 genes are important regulators of HCC cell growth and survival, and may be attractive targets for HCC treatment. Further, SNALP technology for in vivo siRNA delivery may be an effective way to treat human HCC. Future plans include (i) generating more HCC cell lines expressing luciferase for in vivo validation of the therapeutic activity of selected genes;and (ii) understanding the molecular mechanisms underlying growth inhibition and apoptotic induction in Huh7 and HepG2 cells triggered by target gene silencing, using a microarray approach. The ultimate goal of this project is to use siRNA-SNALP therapeutic agents for the treatment of HCC patients. As a first step, Investigative New Drug (IND) applications will be filed for the two most effective siRNAs (COP1 and WEE1) in collaboration with Tekmira Pharmaceuticals and Dr. Avital, Surgery Branch, NCI. Currently we are collecting additional data including toxicology tests, expression pattern in normal tissues, immune-stimulation and immunogenicity, pharmacokinetics, biodistribution, and pharmacodynamics. Full IND-enabling studies are now ongoing for both ApoB and Oncology SNALP programs of Tekmira, including single and multi-dose toxicity and pharmacology testing in mice, rats and Cynomolgus monkeys to support the first human clinical trials of SNALP-based treatments. We examined the transcriptomic and epigenomic changes following ZEB treatment using microarray analysis. To broaden the specificity of the ZEB gene signature, 10 cell lines representing the diversity of human liver cancer, including HCC and ICC, were treated with 100 and 200 M ZEB or vehicle control in triplicate for 7 days prior to microarray analysis. Significant gene lists were computed at P&amp;#8804;0.001 which were universally changed by ZEB in HCC (631) and ICC (451). 64 genes showed significant gene expression changes in both groups. Hierarchical clustering differentiated the cell lines into responders and non-responders following ZEB treatment. Moreover, the ZEB gene signature applied to a data set of 54 human HCCs successfully differentiated the patients according to survival. The antitumor efficacy of ZEB was first evaluated in a model of subcutaneous transplantation of responder (Huh7 and KMCH) and non-responder (WRL68, WITT) cell lines using nude mice (n=3/cell line). Comparison of the kinetics of tumor growth demonstrated the antitumor potency of ZEB only in the responder lines. In contrast, ZEB pretreatment of WRL68 and WITT either did not affect or amplified the rate of tumor growth, consistent with the differential response to ZEB in vitro. To exclude the contribution of the recipient mouse strain, this effect was confirmed by a complimentary analysis of subcutaneous (s.c.) transplantation of ZEB-treated Huh7 and WRL68 cells (n=4/group) into SCID/Beige mice. As a final validation of antitumor efficacy of ZEB in vivo, we used a murine xenograft model. Huh7 cells expressing luciferase (Huh7luc+) were transplanted intrasplenic to immunodeficient SCID/Beige mice. The recipients were randomized based on the intensity of bioluminescence imaging and subjected to ZEB therapy nine days after transplantation. To maximize plasma ZEB exposure, an optimal dosing regimen was selected based on the pharmacokinetics of ZEB and its metabolites in mice using an ADAPT II (BMSR, USC) model. ZEB monotherapy showed a marginal antitumor effect, reducing the mean tumor burden 2-fold 14-days after completion of treatment, which was consistent with a relatively low bioavailability of ZEB in vivo (t of 40 minutes). To increase ZEB bioavalibility, ZEB therapy was combined with a Raloxifene known to inhibit aldehyde oxidase 1 (AOX1), a rate limiting enzyme of ZEB metabolism. The combined therapy significantly increased the antitumor potency of ZEB and caused a 7-fold reduction in tumor mass as compared to a Raloxifene-alone group (P&lt;0.0005). Furthermore, it decreased the frequency and average size of Huh7 pulmonary metastasis (P=0.028). The results suggest that the combination of ZEB with Raloxifene hold clinical promise as an epigenetic anticancer therapy, and that transcriptomic analysis can identify a subset of HCC patients who may benefit the most from epigenetic therapy. Future plans include (i) validating the antitumor efficacy of ZEB in vivo using additional cell lines, (ii) performing survival analysis, (iii) examining the underlying molecular mechanisms of ZEB effects, and (iv) using methylation array (SAM) to investigate the possible usefulness of the ZEB signature to identify HCC patients that might benefit from treatment with inhibitors of DNA methyltransferases.